/**
* Time in milliseconds that connectors will wait for another HTTP request before
* closing the connection. When not set, the connector's container-specific default
* will be used. Use a value of -1 to indicate no (i.e. infinite) timeout.
*/
private Integer connectionTimeout;
I don't know if this is a bug or a feature. However, I definitely feel that this is not reasonable, in tomcat, the default connectionTimeout value is 2000. In spring boot, however, he does not have a default value, causing the system to have a large number of TIME_WAIT. I think a default value should also be given when it is automatically transferred. Not Null or 0.
Comment From: wilkinsona
As described in the comment you have quoted, when server.connection-timeout has not been set, the embedded container is left to use its default timeout for the connector. If you are seeing different behaviour plesse provide a complete sample that reproduces the problem. At this point we don’t even know what version of Boot you are using.
Comment From: hellojukay
@wilkinsona my spring boot version is 1.5.9.RELEASE
Comment From: wilkinsona
Unfortunately, that’s only part of what we need to know. As I said above, if you are seeing different behaviour plesse provide a complete sample that reproduces the problem.
Comment From: hellojukay
@wilkinsona Hello, to reproduces problem is very simple, as long as you create a spring boot program, use the default tomcat configuration, deployment, will your project and then use nginx to agent for this project, of course, want to open the keepalive nginx, ab command is then used to stress test your project, you will find that the spring boot built-in emerged in tomcat time_wait.
Comment From: wilkinsona
A socket being in TIME_WAIT indicates that the socket has been closed and that the TCP stack is waiting for any additional packets. I'm going to close this issue as I don't think there's a problem with Spring Boot. Specifically, I haven't seen anything that suggests that Tomcat's connections are being configured with an indefinite timeout.
Comment From: hellojukay
when the client send header keepalive, spring boot tomcat should not close the socket in 20 seconds, 20 seconds is a default value in tomcat . but spring boot tomcat not include the default value. the value is 6 seconds?
Comment From: wilkinsona
Tomcat's default is 60 seconds. Spring Boot will leave Tomcat to use that default unless you have configured it. I've just verified this behaviour using one of our samples.
If you want someone to spend some more time looking at this, then you'll need to provide a small sample that shows Spring Boot configuring Tomcat with a timeout other than its default.
Comment From: hellojukay
tomcat conf/server.xml, the default timeout is 20000
<Connector executor="tomcatThreadPool"
port="8080" protocol="HTTP/1.1"
connectionTimeout="20000"
redirectPort="8443" />
Comment From: hellojukay
This is my sample
https://github.com/hellojukay/springboot-sample
this is my nginx conf, enable nginx keepalive
#user nobody;
worker_processes 1;
#error_log logs/error.log;
#error_log logs/error.log notice;
#error_log logs/error.log info;
#pid logs/nginx.pid;
events {
worker_connections 1024;
}
http {
include mime.types;
default_type application/octet-stream;
#log_format main '$remote_addr - $remote_user [$time_local] "$request" '
# '$status $body_bytes_sent "$http_referer" '
# '"$http_user_agent" "$http_x_forwarded_for"';
#access_log logs/access.log main;
sendfile on;
#tcp_nopush on;
#keepalive_timeout 0;
keepalive_timeout 120s 120s;
keepalive_requests 10000;
#gzip on;
server {
listen 80;
server_name localhost;
#charset koi8-r;
#access_log logs/host.access.log main;
# location / {
# root html;
# index index.html index.htm;
# }
root /Users/jukay;
#error_page 404 /404.html;
# redirect server error pages to the static page /50x.html
#
error_page 500 502 503 504 /50x.html;
location = /50x.html {
root html;
}
location /hello {
proxy_pass http://127.0.0.1:7777;
}
# proxy the PHP scripts to Apache listening on 127.0.0.1:80
#
#location ~ \.php$ {
# proxy_pass http://127.0.0.1;
#}
# pass the PHP scripts to FastCGI server listening on 127.0.0.1:9000
#
#location ~ \.php$ {
# root html;
# fastcgi_pass 127.0.0.1:9000;
# fastcgi_index index.php;
# fastcgi_param SCRIPT_FILENAME /scripts$fastcgi_script_name;
# include fastcgi_params;
#}
# deny access to .htaccess files, if Apache's document root
# concurs with nginx's one
#
#location ~ /\.ht {
# deny all;
#}
}
# another virtual host using mix of IP-, name-, and port-based configuration
#
#server {
# listen 8000;
# listen somename:8080;
# server_name somename alias another.alias;
# location / {
# root html;
# index index.html index.htm;
# }
#}
# HTTPS server
#
#server {
# listen 443 ssl;
# server_name localhost;
# ssl_certificate cert.pem;
# ssl_certificate_key cert.key;
# ssl_session_cache shared:SSL:1m;
# ssl_session_timeout 5m;
# ssl_ciphers HIGH:!aNULL:!MD5;
# ssl_prefer_server_ciphers on;
# location / {
# root html;
# index index.html index.htm;
# }
#}
include servers/*;
}
this is my case, the 7777 port release connection too fast, i think the connect status should not bee TIME_WAIT,It meant that tomcat quickly turned off the connection without waiting for a default timeout. look at this gif,
Comment From: wilkinsona
This has nothing to do with Spring Boot. You're using curl and it closes all the connections that it has opened when it exits.
Comment From: hellojukay
@wilkinsona no,no,no
you can use ab -k to request , connection TIME_WAIT too.
Comment From: happier2
@wilkinsona no,no,no
you can use ab -k to request , connection TIME_WAIT too.
By default, nginx will use protocal http/1.0 to connect with your spring boot application, which will cause lots of short connection between nginx and tomcat. Lots of short connection(frequently open tcp and close tcp) will produce lots of TIME_WAIT.